You could try to implement middle operation in your process by loading your csv data into an instance of HSQL DB (SQL Database in pure java).
You could execute familiar SQL requests over your csv data after that.
I think, it is better than doing this bicycle once again engine by yourself. Loading your csv into database is more flexible in perpective, when you wanna change your kind of grouping data, or want to add new kind of data representations.
четверг, 4 ноября 2021 г. в 18:40:27 UTC+3, Tatu Saloranta: